Zinc-mediated intramolecular acyl and imino transfer reactions of aryl iodides

Article abstract of DOI:10.1016/j.tetlet.2004.12.034

A method for the coupling of acyl and imino substituents to the sterically encumbered 5-position of a 4-aminoquinazoline was developed. Starting with a 4-amino-5-iodoquinazoline, the method employs a facile intramolecular zinc-mediated transfer from the 4-amino group to the iodo-bearing carbon. The method was found to be effective for a variety of substituents, in particular a pyridyl group required for the synthesis of Pfizer's prostate selective α₁ antagonist candidate for the treatment of benign prostatic hyperplasia, UK-338,003.

Full text of DOI:10.1016/j.tetlet.2004.12.034

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 983–986
Zinc-mediated intramolecular acyl and imino
transfer reactions of aryl iodides
a
Lee T. Boulton, Martin E. Fox, Paul B. Hodgson and Ian C. Lennon
a,
b
a
*
a
Dowpharma, Chirotech Technology Ltd, a subsidiary of The Dow Chemical Company, Unit 321 Cambridge
Science Park, Milton Road, Cambridge CB4 0WG, UK
b
Chemical Research and Development, Pfizer Global Research and Development, Pfizer Ltd, Ramsgate Road,
Sandwich, Kent CT13 9NJ, UK
Received 13October 2004; revised 19 November 2004; accepted 8 December 2004
Abstract—A method for the coupling of acyl and imino substituents to the sterically encumbered 5-position of a 4-aminoquinazoline
was developed. Starting with a 4-amino-5-iodoquinazoline, the method employs a facile intramolecular zinc-mediated transfer from
the 4-amino group to the iodo-bearing carbon. The method was found to be effective for a variety of substituents, in particular a
pyridyl group required for the synthesis of PfizerÕs prostate selective a antagonist candidate for the treatment of benign prostatic
1
hyperplasia, UK-338,003.
Ó 2004 Elsevier Ltd. All rights reserved.
Formation of carbon–carbon bonds in sterically
crowded positions of functionalized molecules is a dif-
ficult challenge in synthetic organic chemistry. The
coupling of a 2-pyridyl substituent to the congested
as Pd- and Ni-catalyzed cross-coupling (Negishi cou-
pling), yet compatible with a wide range of functional-
4
–6
ity.
Owing to their lower toxicity and other
advantageous properties described above, we were at-
tracted to the use of organozinc reagents in this step.
5
-position of the 4-aminoquinazoline 1 was a key step
in a synthesis of PfizerÕs prostate selective a antagonist
1
7
candidate for the treatment of benign prostatic hyper-
Initially, we used 2-pyridylzinc chloride in this reaction
1
–3
plasia, UK-338,003 3 (Scheme 1). The existing meth-
od for this coupling employed a 2-pyridyltin reagent.
However, the toxicity of organotin compounds makes
the use of this reagent unattractive in the synthesis of
a pharmaceutical substance. Organozinc compounds
are very useful and versatile reagents in organic synthe-
sis, being reactive in C–C bond-forming processes such
in place of 2-tributylstannylpyridine (Scheme 2). 2-Pyr-
idylzinc chloride was prepared from 2-bromopyridine
by halogen–metal exchange with isopropylmagnesium
8
–10
chloride,
followed by addition of zinc chloride. The
desired coupling reaction took place, but was accompa-
nied by substantial reduction to 4. We were unable to
achieve more than a 1:1 ratio of 2–4 in this reaction.
NHSO2Me
Pd(PPh3)4,
CuI, LiCl
MeO
MeO
N
Cl
MeO
MeO
N
Cl
MeO
MeO
N
N
SnBu3
N
N
N
+
N
NH2
I
NH2
NH2
N
N
1
2
3
Scheme 1. Synthesis of UK-338,003 employing 2-(tributylstannyl)pyridine.
Keywords: Organozinc reagent; Cross-coupling reactions; Negishi coupling; Heterocyclic chemistry; Aromatic substitution.
*
Corresponding author. Tel.: +44 0 12237280 38 ; fax: +44 0 1223506701; e-mail: mfox@dow.com
0
040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.12.034

984
L. T. Boulton et al. / Tetrahedron Letters 46 (2005) 983–986
2
-PyZnCl, Pd(OAc)2 (cat.),
MeO
MeO
N
Cl
MeO
MeO
N
Cl
MeO
MeO
N
Cl
PPh3 (cat.), THF, reflux
+
N
N
N
or Zn, 2-iodopyridine, Pd(OAc)2 (cat.),
NH2
I
NH2
o
NH2
PPh3 (cat.), DMF, 50 C
N
1
2
4
Scheme 2. Negishi couplings of iodide 1.
Therefore we investigated the possibility of inverting the
nucleophilic and electrophilic partners in the reaction
and employing the zinc reagent formed by iodine–metal
exchange of 1. The iodide 1 was treated with activated
zinc in DMF, followed by 2-iodopyridine and a palla-
dium catalyst. Only the reduced compound 4 was pro-
duced. We reasoned that metallation of the iodide 1
had occurred, but that due to the proximity of the acidic
N–H protons, intramolecular protonation of the
organozinc intermediate had occurred, giving rise to
the observed reduced product 4.
of the sparingly soluble product rather than the occur-
rence of side reactions. The N-diacetyl compound 7
underwent an analogous reaction, giving rise to the
methyl ketone 8. This compound exists, in CDCl , as
3
a 3:1 mixture of cyclic hemi-aminal 9 and ketone 8
species.
1
1
It was clear the use of carbonyl-based N-protecting
groups was unlikely to allow formation of a stable
organozinc halide. Nevertheless, the acyl migration
reaction was a synthetically interesting transformation.
The migration of functionalized acyl groups was also
successful (Scheme 4). Starting with the imide 10, after
zinc-mediated migration of one of the 5-benzoyloxyvale-
ryl groups, the benzoyl and remaining N-acyl groups
were cleaved with methanolic potassium carbonate to
To prevent quenching of the zinc reagent, we protected
the free NH group as a phthalimide (Scheme 3), a
group previously used to protect amino groups in
2
1
2
organozinc reagents. To our surprise, on treatment
of 5 with activated zinc at room temperature, the iodide
underwent not reductive de-iodination as with the
unprotected amine 1, but smooth conversion to the 8-
membered lactam 6, the product of migration of one
of the phthalimide carbonyls to the iodine-bearing car-
bon, presumably by nucleophilic attack of the inter-
1
. Zn, DMF
MeO
MeO
N
N
Cl
O
MeO
MeO
N
Cl
2. K CO ,
2
3
N
MeOH
N
I
O
NH2
1
3
O
mediate organozinc reagent in a Barbier-like reaction.
64%
This reaction was remarkably facile in occurring at
room temperature. We attempted to capture the pre-
sumed intermediate organozinc species by addition of
2-iodopyridine, palladium acetate and triphenylphos-
phine, but the lactam 6 was the only product. The low
yield (34%) appears to reflect the difficulty in isolation
OH
OBz
OBz
10
11
Scheme 4. Migration of functionalized acyl group.
MeO
N
N
Cl
O
MeO
MeO
N
Cl
N
N
MeO
Zn, DMF, rt
4%
I
O
O
NH
O
Phthaloyl dichloride,
3
Et3N, chlorobenzene,
7
5 o
C
75%
5
6
MeO
MeO
N
Cl
N
MeO
MeO
N
Cl
Cl
I
NH2
NaH, AcCl,
THF, rt
N
O
HN
N
MeO
MeO
N
N
Cl
O
O
1
60%
8
N
Zn, DMF, rt
0%
+
I
9
O
MeO
MeO
7
N
N
O
HO
9
Scheme 3. Zinc-mediated acyl migration reactions of quinazoline 1.

L. T. Boulton et al. / Tetrahedron Letters 46 (2005) 983–986
985
DMF, (COCl)2,
CH2Cl2
MeO
MeO
N
Cl
MeO
MeO
N
N
Cl
MeO
MeO
N
Cl
o
Zn, DMF, rt-40 C
N
N
N
I
NH2
I
N
H
90%
56%
Me2N
13
NMe2
1
12
Scheme 5. Zinc-mediated transfer reaction of formamidine 12.
give the 5-hydroxyvaleryl ketone 11. The intermolecular
version of the acyl transfer reaction would also be of
interest, but would not provide a solution to the current
problem and was not investigated. In the search for
alternative N-protecting groups, we tried a formamidine
6). The 4-aminoquinazoline 1 was converted to the
8
dichloride 14 using a modified Sandmeyer reaction.
1
Chlorination was accompanied by formation of the 4-
quinazolone 15. The crude product mixture was reacted
with phosphoryl chloride to convert the quinazolone 15
to the dichloride 14. Reaction of the dichloride 14 with
2-aminopyridine was selective for substitution at the 4-
position, thus providing the required aminopyridine
16. In order to effect pyridyl transfer, activation of the
1
4–17
(
Scheme 5).
Treatment of the iodide 12 with acti-
vated zinc led to the corresponding C–C bond-forming
reaction with the amidine carbon, to give the cyclic ami-
nal 13, a derivative of the 5-formylquinazoline.
2-aminopyridine 16 to an imino species analogous to
The preceding examples suggested that the N–C transfer
reaction in this system was likely to be a general process.
Therefore we decided to attempt to employ this method
for introduction of the desired 2-pyridyl group. In order
to prepare a suitable precursor 16, coupling of the 2-pyr-
idyl group to the amino group of 1 was required. How-
ever, we had previously found the amino group of 1 to
be only weakly nucleophilic, requiring forcing reaction
conditions to achieve acetylation for example. We at-
tempted to couple this group by reaction of the amino-
quinazoline 1 with 2-fluoropyridine using sodium
hydride as a base. However, no reaction took place at
room temperature, and on heating, de-iodination took
place. Since aromatic nucleophilic substitution required
conditions too forcing for the iodide to tolerate, and the
alternative of a palladium-catalyzed coupling was also
likely to be incompatible with this functionality, we
adopted a more circuitous route in which the nucleo-
philic and electrophilic partners were inverted (Scheme
the formamidine 12 was required. This was achieved
by N-acylation of the pyridine. Reaction of the 2-amino-
pyridine 16 with di-tert-butyl dicarbonate led to the for-
mation of the bright yellow pyridimine 17. Isolation of
the Boc pyridimine 17 was not possible, but the com-
pound was stable for hours in solution at room temper-
ature, allowing its identity to be confirmed by NMR.
On storage, the pyridimine 17 rearranges to the N-Boc
aminopyridine 18 (Fig. 1). When acetic anhydride or
1
9
MeO
MeO
N
N
Cl
N
N
I
O
O
18
Figure 1. Rearrangement product of pyridimine 17.
CuCl2, isobutyl nitrite,
MeCN, reflux
2
-aminopyridine,
MeO
MeO
N
Cl
MeO
MeO
N
Cl
MeO
MeO
N
Cl
N
NaH, THF, rt
N
N
N
92%
I
NH2
I
Cl
I
HN
1
14
POCl , EtiPr N
16
3
2
MeCN, reflux
+
6
1
5% from
over 2 steps
MeO
MeO
N
O
Cl
NH
Boc O,
DMF
2
I
15
MeO
MeO
N
N
Cl
MeO
MeO
N
Cl
MeO
MeO
N
Cl
N
N
N
CF3COOH
Zn
N
I
NH
O
NH2
54% over 3
steps from 16
without isolation
N
O
N
O
O
2
19
17
Scheme 6. Zinc-mediated pyridyl transfer.

986
L. T. Boulton et al. / Tetrahedron Letters 46 (2005) 983–986
tosyl chloride were used instead of di-tert-butyl dicar-
bonate, a yellow colour also resulted, signifying forma-
tion of the corresponding pyridimines, but these were
much less stable than the N-Boc species 17.
Crook, R. J.; Pettman, A. J.; Walton, R. Chem. Abstr.
003, 138, 170247.
. Huo, S. Org. Lett. 2003, 5, 423–425.
. Knochel, P.; Millot, N.; Rodriguez, A. L.; Tucker, C. E.
Org. React. 2001, 58, 417.
2
4
5
6
. Boudier, A.; Bromm, L. O.; Lotz, M.; Knochel, P. Angew.
Chem., Int. Ed. 2000, 39, 4414–4435.
. Prasad, A. S.; Stevenson, T. M.; Citineni, J. R.; Nyzam,
V.; Knochel, P. Tetrahedron 1997, 53, 7237–7254.
We were pleased to find that on addition of a pre-
formed solution of the N-Boc pyridimine 17 in DMF
to a suspension of zinc in DMF pre-activated with tri-
methylsilyl chloride and 1,2-dibromoethane that smooth
migration of the pyridyl group occurred to give the spi-
rocyclic dihydropyridine 19. Upon treatment with trifluo-
roacetic acid, the spirocycle 19 collapsed to the desired
pyridine 2. The steps from the 2-aminopyridine 17 could
be carried out without purification of the spirocycle 19,
giving the pyridine 2 in 54% overall yield. Thus, we were
able to achieve introduction of a variety of acyl substit-
uents and a 2-pyridyl group to the 5-position of the qui-
nazoline 1. We believe that the transfer reaction
represents a valuable method for introduction of a range
of substituents into sterically crowded and highly func-
tionalized aryl systems.
7
8. Krasovsky, A.; Knochel, P. Angew. Chem., Int. Ed. 2004,
43, 3333–3336.
9. Abarbri, M.; Thibonnet, J.; B e´ rillon, L.; Dehmel, F.;
Rottl a¨ nder, M.; Knochel, P. J. Org. Chem. 2000, 65, 4618–
4
0. Tr e´ court, F.; Breton, G.; Bonnet, V.; Mongin, F.; Mars-
634.
1
ais, F.; Qu e´ guiner, G. Tetrahedron Lett. 1999, 40, 4339–
4
342.
1. Knochel, P.; Yeh, M. C. P.; Berk, S. C.; Talbert, J. J. Org.
Chem. 1988, 53, 2390–2392.
2. Lipshutz, B. H.; Blomgren, P. A. J. Am. Chem. Soc. 1999,
121, 5819–5820.
1
1
13. Blomberg, C. The Barbier Reaction and Related One-step
Processes; Springer: Berlin, 1993.
1
1
1
1
1
4. Staubitz, A.; Dohle, W.; Knochel, P. Synthesis 2003, 233–
2
5. Varchi, G.; Jensen, A. E.; Dohle, W.; Ricci, A.; Cahiez,
G.; Knochel, P. Synlett 2001, 477–480.
6. Meyers, A. I.; Milot, G. J. Am. Chem. Soc. 1993, 115,
42.
Acknowledgements
This work was carried out in collaboration with Pfizer
UK Ltd, Chemical Research and Development, Sand-
wich, Kent, CT139NJ, UK.
6
652–6660.
7. Gottlieb, L.; Meyers, A. I. Tetrahedron Lett. 1990, 31,
723–4726.
8. Doyle, M. P.; Siegfried, B.; Dellaria, J. F. J. Org. Chem.
977, 42, 2426–2431.
4
1
References and notes
19. All new compounds gave satisfactory spectral and ana-
lytical data. H NMR spectrum of pyridine 16 (400 MHz,
1
1
2
. Fox, D. N. A. Patent application WO 9830560, 1998.
Chem. Abstr. 1998, 129, 136176.
. Ahman, J. B.; Hodgson, P. B.; Lewandowski, S. J.;
Walton, R. Patent application WO 2002094815, 2002;
Ahman, J. B.; Hodgson, P. B.; Lewandowski, S. J.;
Walton, R. Chem. Abstr. 2003, 138, 4608.
acetone-d ) d ppm 8.56 (1H, d, J = 8.0), 8.40 (1H, d,
6
J = 5.0), 7.97 (1H, s), 7.91 (1H, td, J = 7.8, 1.0), 7.29 (1H,
s), 7.18 (1H, dd, J 7.6, 5.0), 4.10 (3H, s) and 3.93 (3H, s)
and H NMR spectrum of pyridimine 17 recorded in situ
1
(400 MHz, acetone-d ) d ppm 7.84 (1H, dt, J = 6.8, 1.0),
6
7.47 (1H, ddd, J = 9.1, 6.4, 1.0), 7.30 (1H, d, J = 9.1) 7.02
(1H, s), 6.44 (1H, td, J = 6.7, 1.0), 3.91 (3H, s), 3.68 (3H, s)
and 1.41 (9H, s).
3
. Boulton, L. T.; Crook, R. J.; Pettman, A. J.; Walton, R.
Patent application WO 2003011829, 2003; Boulton, L. T.;